trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999934373154245,
  "global_step": 8571,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "learning_rate": 1.9420185252014917e-05,
      "loss": 0.5431,
      "step": 500
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.8217250090220138e-05,
      "loss": 0.3568,
      "step": 1000
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.701431492842536e-05,
      "loss": 0.2985,
      "step": 1500
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.581137976663058e-05,
      "loss": 0.2649,
      "step": 2000
    },
    {
      "epoch": 0.88,
      "learning_rate": 1.4608444604835801e-05,
      "loss": 0.241,
      "step": 2500
    },
    {
      "epoch": 1.05,
      "learning_rate": 1.340550944304102e-05,
      "loss": 0.209,
      "step": 3000
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.2202574281246243e-05,
      "loss": 0.1683,
      "step": 3500
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.0999639119451462e-05,
      "loss": 0.1611,
      "step": 4000
    },
    {
      "epoch": 1.58,
      "learning_rate": 9.796703957656684e-06,
      "loss": 0.1536,
      "step": 4500
    },
    {
      "epoch": 1.75,
      "learning_rate": 8.593768795861904e-06,
      "loss": 0.1457,
      "step": 5000
    },
    {
      "epoch": 1.93,
      "learning_rate": 7.390833634067124e-06,
      "loss": 0.141,
      "step": 5500
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.187898472272345e-06,
      "loss": 0.1142,
      "step": 6000
    },
    {
      "epoch": 2.28,
      "learning_rate": 4.9849633104775655e-06,
      "loss": 0.0962,
      "step": 6500
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.7820281486827865e-06,
      "loss": 0.0946,
      "step": 7000
    },
    {
      "epoch": 2.63,
      "learning_rate": 2.579092986888007e-06,
      "loss": 0.091,
      "step": 7500
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3761578250932275e-06,
      "loss": 0.0891,
      "step": 8000
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.7322266329844823e-07,
      "loss": 0.0877,
      "step": 8500
    },
    {
      "epoch": 3.0,
      "step": 8571,
      "total_flos": 1.8800594957369344e+16,
      "train_loss": 0.19064752925667625,
      "train_runtime": 224598.7373,
      "train_samples_per_second": 19.539,
      "train_steps_per_second": 0.038
    }
  ],
  "max_steps": 8571,
  "num_train_epochs": 3,
  "total_flos": 1.8800594957369344e+16,
  "trial_name": null,
  "trial_params": null
}